Glycosphingolipids (GSLs) have been implicated in various stages of HIV-1 infection [1–3]. One particular GSL that has been long associated with the mechanism of HIV-1 infection is globotriaosylceramide (Gb3) also known as CD77 and the Pk blood group antigen [4–9]. Recent studies using conditions closer to the natural infection pathophysiology of HIV have shown that cell surface expression of Gb3 may provide natural resistance against HIV-1 infection .
The first study suggesting the protective role of Gb3 came from the use of a semi-synthetic, water-soluble analogue of Gb3. Preincubation of the virus with adamantylGb3 inhibited HIV-1 infection independent of strains of virus and drug-resistant status and prevented fusion of HIV-1 and HIV-2 to target cells . Additional reports supported a role for Gb3 as a resistance factor to HIV-1 [12–15]. Although it appears clear that soluble Gb3 mimetics may have efficacy in treatment/prevention of HIV infection [4,5,11–13], a pharmacological approach to increase Gb3 on CD4+ T-cells  requires these cells to express Gb3. However, human CD4+ T-cell Gb3 expression has never been carefully evaluated. Our studies specifically address this question. We show that human CD4+ T-cells are not capable of significant Gb3 expression; thus, a treatment/prevention strategy for HIV infection must focus on development of soluble Gb3 mimetics.
We used a natural ligand for Gb3, the pentameric B subunit of verotoxin 1 (VT1B) conjugated to the flurochrome Alexa488 to examine cell-surface Gb3 expression by fluorescence-activated cell sorting analysis . Expression of Gb3 on unstimulated, circulating human peripheral blood mononuclear cells (PBMCs) was negligible (<1% of total cells). Among these cells, the majority of Gb3-expressing cells were positive for CD4, representing T-cells, natural killer (NK)-cells, or monocytes with the expression level of individual cell types CD4>CD16>CD14>CD56>CD8>CD19 (Fig. 1a). Expression of cell surface-associated Gb3 in activated T-cells was examined after stimulation with phytohemagglutinin (PHA), PHA/IL-2, anti-CD3, anti-CD3/IL-2, or phorbol 12-myristate 13-acetate/ionomycin. None of the activation protocols increased Gb3 expression significantly. The best result was achieved using PHA/IL-2 but, after 9 days of stimulation, Gb3 only increased from 0.5 to 3.2% of total cells (Fig. 1b).
To determine the phenotype of Gb3-expressing PBMCs, four-color flow cytometry analysis was used. The cell-specific markers used with CD4 and VT1B included combinations of CD25, FOXP3, CD56, and CD1d tetramer. Results showed a small proportion of Gb3+, activated CD4+ CD25+ T-cells that also expressed FOXP3 and CD56 (Fig. 1c and d). There were no CD4+ T-cells expressing Gb3 that bound the CD1d tetramer, indicating Gb3 is likely not expressed on NKT cells.
The low percentage of Gb3-expressing cells in various subsets of stimulated CD4+ T-cells made characterization of these cells difficult. Therefore, a strategy was developed that used fluorescein isothiocyanate (FITC)-labeled VT1B and anti-FITC conjugated to immunomagnetic beads to positively select Gb3-expressing cells following stimulation. Most of the enriched Gb3-expressing cells expressed CD4 (Fig. 1e). Further analysis revealed that, in this particular experiment, 15.4% of these cells were coexpressing CD25 and CD56, whereas 77.0% were associated with CD25 and FOXP3 (Fig. 1e). Thus, we confirmed the previous characterization that most of these cells were positive for CD25 and FOXP3, whereas some cells expressed CD56.
Thin layer chromatography using verotoxin B overlay indicated that CD4+ T-cells lacking cell surface-expressed Gb3 also did not contain masked or intracellular Gb3 (data not shown). To confirm this finding, 14C-galactose metabolically labeled CD4+ T-cells were purified, and GSLs from only the cells that lacked surface expression of Gb3 were processed (Fig. 1f). GSL bands were detected in lanes 4 and 5, which consisted of GSLs extracted from anti-CD3/IL-2-stimulated CD4+ T-cells and PHA/IL-2-stimulated CD4+ T-cells, respectively. Despite the presence of GlcCer, LacCer, and GM3 in these lanes, Gb3 and Gb4 were undetectable in these cell extracts. The result for the control lane, lane 2, which consisted of GSLs extracted from the total PHA/IL-2-stimulated PBMCs, was consistent with previous published findings  and likely represents non-CD4+ T-cells.
The current study is the first to examine expression of Gb3 in human CD4+ T-cells. Using an optimal detection method for Gb3 that uses flow cytometry analysis , we showed very low levels of cell surface-expressed Gb3 on either resting or activated human PBMCs (Fig. 1a). We showed that in the activated CD4+ T-cells, only a very small subpopulation of human CD4+ T-cells upregulate surface-expressed Gb3. These T-cells had phenotypic characteristics of T-regulatory (Treg) and natural killer (NK) cells. These results were confirmed after enrichment of Gb3-expressing cells. Thus, we can conclude that small populations of Tregs or NK-like cells may express Gb3 but not the majority of CD4+ T-cells.
Furthermore, we showed that CD4+ T-cells lacking cell-surface expression of Gb3 upon stimulation did not express masked or intracellular Gb3. Our conclusion from these findings is that human CD4+ T-cells express negligible levels of cell-surface or intracellular Gb3 in either a resting or activated state.
HIV-1 infects human CD4+ T-cells but does not infect any other T-cells in mammals, with the exception of certain nonhuman primates, particularly the chimpanzee . Indeed, it would be interesting to examine other mammalian T-cells for expression of Gb3 and to determine whether a lack of infection of these cells is related to the cell-surface expression of this GSL. In previous work, we have shown Gb3 expression to provide natural resistance for HIV infection . We hypothesize that human CD4+ T-cells are highly permissive to HIV infection because, in part, they cannot express the natural resistance factor, Gb3, under any conditions.
As the cell surface expression of Gb3 was detected only in a very small percentage of CD4+ T-cells after in-vitro stimulation, the possible role of cholesterol-masked or intracellular expression of Gb3 was examined. Using metabolic labeling of GSLs with 14C-galactose on purified, activated CD4+ T-cells that lacked surface-associated Gb3, we showed that in-vitro stimulation did not induce Gb3 synthesis in these cells, confirming that T-cells cannot make Gb3.
The current study was undertaken in order to understand whether T-cell expression of Gb3 occurs and, thus, could be beneficial in the prevention of HIV-1 infection. Our findings that little if any Gb3 is expressed in human CD4+T-cells, even under conditions of activation, indicate that T-cells would not be pharmacologic targets for increasing Gb3 expression in vivo, with the aim to inhibit HIV-1 infection. Indeed, we found that 1-deoxygalactojirimycin (DGJ), shown in cell lines to upregulate Gb3 and increase cellular resistance to HIV-1 infection , failed to affect HIV-1 infection using PBMCs (data not shown). Nonetheless, the role of Gb3 as a resistance factor against HIV-1 infection should not be minimized as this GSL still holds potential for future therapeutic and prevention strategies against HIV/AIDS as a soluble analogue mimetic [4–6,13].
Conflicts of interest
This work was funded by operating grants from the Ontario HIV Treatment Network (OHTN; D.R.B. and C.A.L.) and the Canadian Foundation for AIDS Research (CANFAR; C.A.L. and D.R.B.), as well as a studentship from the Natural Sciences and Engineering Research Council (NSERC) of Canada (M.K.). The authors thank Dr Peter van den Elzen (University of British Columbia) for his kind gift of CD1d tetramer used in these studies. Drs A. Lingwood and D. R. Branch are founders and stock holders in ViroCarb Inc. There are no other conflicts of interest.
1. Harouse JM, Laughlin MA, Pletcher C, Friedman HM, Gonzalez-Scarano F. Entry of human immunodeficiency virus-1 into glial cells proceeds via an alternate, efficient pathway
. J Leukoc Biol
2. Yahi N, Baghdiguian S, Moreau H, Fantini J. Galactosyl ceramide (or a closely related molecule) is the receptor for human immunodeficiency virus type 1 on human colon epithelial HT29 cells
. J Virol
3. Waheed AA, Freed EO. Lipids and membrane microdomains in HIV-1 replication
. Virus Res
4. Lingwood CA, Branch DR. The role of glycosphingolipids in HIV/AIDS
. Discov Med
5. Lingwood CA, Branch DR. Glycosphingolipids in HIV/AIDS: the potential therapeutic application.
Understanding HIV/AIDS management and care–pandemic approaches in the 21st Century. In: Kasenga FH, editor. InTech AIDS Book 4.2011. pp. 239–280.
6. Branch DR. Blood groups and susceptibility to virus infection: new developments
. Curr Opin Hematol
7. Fantini J, Maresca M, Hammache D, Yahi N, Delezay O. Glycosphingolipid (GSL) microdomains as attachment platforms for host pathogens and their toxins on intestinal epithelial cells: activation of signal transduction pathways and perturbations of intestinal absorption and secretion
. Glycoconj J
8. Hammache D, Yahi N, Maresca M, Pieroni G, Fantini J. Human erythrocyte glycosphingolipids as alternative cofactors for human immunodeficiency virus type 1 (HIV-1) entry: evidence for CD4-induced interactions between HIV-1 gp120 and reconstituted membrane microdomains of glycosphingolipids (Gb3 and GM3)
. J Virol
9. Puri A, Hug P, Jernigan K, Barchi J, Kim H, Hamilton J, et al. The neutral glycosphingolipid globotriaosylceramide promotes fusion mediated by a CD4-dependent CXCR4-utilizing HIV type 1 envelope glycoprotein
. Proc Natl Acad Sci U S A
10. Lund N, Olsson ML, Ramkumar S, Sakac D, Yahalom V, Levene C, et al. The human Pk histo-blood group antigen provides protection against HIV-1 infection
11. Lund N, Branch DR, Mylvaganam M, Chark D, Ma X-Z, Sakac D, et al. A novel soluble mimic of the glycolipid, globotriaosyl ceramide inhibits HIV infection
12. Harrison AL, Olsson ML, Jones RB, Ramkumar S, Sakac D, Binnington B, et al. A synthetic globotriaosylceramide analogue inhibits HIV-1 infection in vitro by two mechanisms
. Glycoconj J
13. Rosa Borges A, Wieczorek L, Johnson B, Benesi AJ, Brown BK, Kensinger RD, et al. Multivalent dendrimeric compounds containing carbohydrates expressed on immune cells inhibit infection by primary isolates of HIV-1
14. Lund N, Branch DR, Sakac D, Lingwood CA, Siatskas C, Robinson CJ, et al. Lack of susceptibility of cells from patients with Fabry disease to productive infection with R5 human immunodeficiency virus
15. Ramkumar S, Sakac D, Binnington B, Branch DR, Lingwood CA. Induction of HIV-1 resistance: cell susceptibility to infection is an inverse function of globotriaosyl ceramide levels
16. Kim M, Binnington B, Sakac D, Fernandes KR, Shi SH, Lingwood CA, et al. Comparison of detection methods for cell surface globotriaosylceramide
. J Immunol Methods
17. Fultz PN. Nonhuman primate models for AIDS
. Clin Infect Dis
1993; 17 (Suppl 1):S230–S235.